Unsupervised Dense Retrieval with Relevance-Aware Contrastive Pre-Training

Yibin Lei, Liang Ding, Yu Cao, Changtong Zan, Andrew Yates, Dacheng Tao


Abstract
Dense retrievers have achieved impressive performance, but their demand for abundant training data limits their application scenarios. Contrastive pre-training, which constructs pseudo-positive examples from unlabeled data, has shown great potential to solve this problem. However, the pseudo-positive examples crafted by data augmentations can be irrelevant. To this end, we propose relevance-aware contrastive learning. It takes the intermediate-trained model itself as an imperfect oracle to estimate the relevance of positive pairs and adaptively weighs the contrastive loss of different pairs according to the estimated relevance. Our method consistently improves the SOTA unsupervised Contriever model on the BEIR and open-domain QA retrieval benchmarks. Further exploration shows that our method can not only beat BM25 after further pre-training on the target corpus but also serves as a good few-shot learner. Our code is publicly available at https://github.com/Yibin-Lei/ReContriever.
Anthology ID:
2023.findings-acl.695
Volume:
Findings of the Association for Computational Linguistics: ACL 2023
Month:
July
Year:
2023
Address:
Toronto, Canada
Editors:
Anna Rogers, Jordan Boyd-Graber, Naoaki Okazaki
Venue:
Findings
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
10932–10940
Language:
URL:
https://aclanthology.org/2023.findings-acl.695
DOI:
10.18653/v1/2023.findings-acl.695
Bibkey:
Cite (ACL):
Yibin Lei, Liang Ding, Yu Cao, Changtong Zan, Andrew Yates, and Dacheng Tao. 2023. Unsupervised Dense Retrieval with Relevance-Aware Contrastive Pre-Training. In Findings of the Association for Computational Linguistics: ACL 2023, pages 10932–10940, Toronto, Canada. Association for Computational Linguistics.
Cite (Informal):
Unsupervised Dense Retrieval with Relevance-Aware Contrastive Pre-Training (Lei et al., Findings 2023)
Copy Citation:
PDF:
https://aclanthology.org/2023.findings-acl.695.pdf